Valve's L4D2 Linux Presentation Slides

[elanthis]

Now, it may turn out that Linux is faster then Windows. Its possible [and frankly, it *should* be given how much Windows does in the background]. But based on years of data, I find it very unlikely OGL would run faster then DX, at least on a Windows based OS.

Linux as a kernel is generally faster than Windows in my experience. I wouldn't doubt that one.

I already pointed out that yes, the draw call overhead in OGL is less than D3D in Windows, because of how D3D runs almost entirely in a separate process (so there's extra context switches and IPC for every D3D call) for security and stability reasons while the GL drivers do not.

It's like comparing OpenGL on classic Mesa vs OpenGL on Gallium. You get totally different performance numbers even though they're both using the extra same frontend Mesa OpenGL code. You even see classic Mesa running faster, or with more features, or with more stability in some cases, even though Gallium is generally considered the superior backend.

Where things go in D3D's favor are the API requiring less overhead (and yes, this _is real_, which is specifically why NVIDIA created all those fancy NV_ extensions for bindless graphics and direct state access that aren't in OpenGL proper -- they really do make a difference!) and the case that the D3D drivers on Windows are usually of higher quality than their OGL counterparts simply because the drivers are tested and vetted by far more apps.

[gamerk2]

And who cares about stupid trolls thoughts? If there's a toy OS it's Windows and this was proven many times (nobody serious puts GUI into ring 0!). It was also proven you're a dumb troll:

http://phoronix.com/forums/showthrea...427#post280427

The sad thing you're still trolling even after being proven wrong.

Well, putting the GUI in ring 0 does make some sense for a closed source OS. MS treats the GUI like a key OS component, not a useland application.

[elanthis]

Well, putting the GUI in ring 0 does make some sense for a closed source OS. MS treats the GUI like a key OS component, not a useland application.

Not sure what being closed source has to do with a technical design decision. In any case, Linux used to (still does?) run X11 as root, which is not a whole lot better than being in the kernel.

[johnc]

Not sure what being closed source has to do with a technical design decision. In any case, Linux used to (still does?) run X11 as root, which is not a whole lot better than being in the kernel.

Still does. I'm not sure why though.

[elanthis]

Still does. I'm not sure why though.

Lame. I remember rootless X being done a whole ago. Should be standard by now.

[kraftman]

Linking to your own baseless post does not show proof.

It does show me that you are now the newest person to be added to my ignore list with the likes of Q and crazycheese and the other loonies that Phoronix for some reason attracts. Bye!

Farewell trolly. There's a proof in my post that you're completely wrong as usual M$ boy.

[kraftman]

On a different OS in a different state of development. Do we know for sure every D3D graphical feature has been implemented? How much of the increase is due to the different OS? Choice of scheduler makes a difference? AMD/NVIDIA?

So yeah, I never take the results of one benchmark as meaning anything. Doing so is just silly.

Now, it may turn out that Linux is faster then Windows. Its possible [and frankly, it *should* be given how much Windows does in the background]. But based on years of data, I find it very unlikely OGL would run faster then DX, at least on a Windows based OS.

That's interesting reading, but the facts stands on the Linux and OpenGL side. L4D2 was for years optimized for Windows and after some little tweaking it runs faster with OpenGL on Windows and it runs even faster on Linux. New Windows box doesn't do too much in the background, but if it does it's its problem. It's Unity that slows L4D2 on Linux, so in a fair benchmark it will run even faster. Unity doesn't suspend compositions for now, so Windows had advantage.

[kraftman]

Linux as a kernel is generally faster than Windows in my experience. I wouldn't doubt that one.

I already pointed out that yes, the draw call overhead in OGL is less than D3D in Windows, because of how D3D runs almost entirely in a separate process (so there's extra context switches and IPC for every D3D call) for security and stability reasons while the GL drivers do not.

It's just a bullshit or PR, because Windows is not more secure and it's not more stable, so running anything in a different process doesn't help it much. It's just a moron made OS.

[kraftman]

Not sure what being closed source has to do with a technical design decision. In any case, Linux used to (still does?) run X11 as root, which is not a whole lot better than being in the kernel.

It's a whole better, because you can run Linux without X and it won't introduce such stability problems like GUI running in the kernel. To add some more facts to the discussion:

http://en.wikipedia.org/wiki/Compari...oft_Windows.29

It's d3d 10 that copied OpenGL solution and become faster than it's predecessors (but it seems it's still slower than OpenGL). Windows approach has nothing to stability and security, but it was just a design mistake.

Since OpenGL's IHV drivers have a user-mode component to them, IHVs have the ability to implement marshalling, thus improving performance. There is still kernel-mode switching, but the theoretical maximum number of switches under OpenGL implementations is simply equal to the Direct3D standard behavior.

Direct3D 10, the release included with Windows Vista,^[9] allows portions of drivers to run in user-mode, making it possible for IHVs to implement marshalling, thus bringing the two back into relative performance parity.

[johnc]

That's interesting reading, but the facts stands on the Linux and OpenGL side. L4D2 was for years optimized for Windows and after some little tweaking it runs faster with OpenGL on Windows and it runs even faster on Linux. New Windows box doesn't do too much in the background, but if it does it's its problem. It's Unity that slows L4D2 on Linux, so in a fair benchmark it will run even faster. Unity doesn't suspend compositions for now, so Windows had advantage.

I'm somewhat curious about the Unity thing myself but I'm thinking they ran it in a window (since the resolution was 1280x1024) where maybe you wouldn't get the performance hit.